Predicting NIV failure in hypoxemic patients: the HACOR score
ICM ARTICLE REVIEW
Non-invasive ventilation (NIV) is applied worldwide to patients with hypoxemic respiratory failure. It is often applied as an attempt to avoid invasive mechanical ventilation. However, the application of NIV is often ad hoc and non-evidence based. Duan et al. have developed a scoring system which accurately predicts patients that would be at risk of NIV failure such that the clinician can plan for the decision to implement invasive mechanical ventilation.
Methods:
Duan et al utilised a test cohort of 449 patients and performed a univariate analysis to determine variables associated with NIV failure at one hour of NIV. They subsequently developed a regression model by entering variables from step 1 that showed a p<0.2 into a stepwise multivariate logistic regression analysis. They subsequently developed a risk scale for each variable. This scale was then applied to a prospective validation cohort.
Results:
1. Heart rate, Acidosis (pH), Consciousness (GCS), Oxygenation, and Respiratory rate (HACOR) were independent predictors of NIV failure in the test cohort.
2. HACOR score is out of 25 with differential weighting of each scale:
3. At 1 hour of NIV, odds ratio of NIV failure is 1.73 for every 1-point increase in HACOR score of test cohort.
4. Patients with NIV failure show a higher HACOR score at 1, 12, 24, and 48 hours of NIV.
5. HACOR score improves in patients with NIV success and remains unaltered in patients with NIV failure.
6. The diagnostic accuracy for NIV failure of a HACOR score above 5 at 1 hour of NIV was 81.8% (test cohort) and 86% (validation cohort).
7. This remained above 80% regardless of NIV duration, diagnosis, age, or disease severity (APACHE 2 score).
8. Patients who failed NIV and were intubated early (within 12 hours) had a HACOR > 5 at NIV initiation and 1 hour NIV than those intubated later (after 12 hours of NIV). Interestingly, early intubation in this study had a significantly lower mortality than late intubation in those with NIV failure.
Conclusions:
1. HACOR is a potentially useful bedside tool for the prediction of NIV failure.
2. HACOR score accurately predicts NIV failure in patients with hypoxemic respiratory failure in this single centre study.
3. A HACOR score >5 at 1hour of NIV highlights patients with a >80% risk of NIV failure regardless of diagnosis, age, and disease severity.
4. The authors imply the utility of HACOR to assess the need for early to improve mortality.
Future clinical trials are needed to define the applicability of the HACOR score and provide further validation of these study findings in alternative settings.
This article review was submitted by EJRC member Dr Brijesh V Patel on behalf of the ESICM Communications Committee.
Reference
Jun Duan, Xiaoli Han, Linfu Bai, Lintong Zhou and Shicong Huang. Assessment of heart rate, acidosis, consciousness, oxygenation, and respiratory rate to predict noninvasive ventilation failure in hypoxemic patients. Intensive Care Med
DOI 10.1007/s00134-016-4601-3
